A latest artificial intelligence platform is being used by Royal Dutch Shell PLC, to operate its predictive maintenance efforts and extend AI-charged applications throughout the company.

The objective is to have more wide access to machine learning and other devices throughout the Shell, expanding and integrating AI applications at broad level, Yuri Sebregts, technology’s executive vice president and technology’s leading officer, claimed.

Shell will employ Microsoft Corporation’s Azure and C3 IoT technology to forecast when valves, compressors and related tools need maintenance; to assist in accurate drilling through deposits of shale; and to enhance the security given to staff members and customers. Shell is also employing Artificial Intelligence (AI) equipment from Bonsai – a corporation, bought by Microsoft in the beginning of this year, which provides software for assisting autonomous operation of computers.

Shell claims that AI-powered horizontal drilling, by which geologists are able to draw an exact path for the well using the real-time data supplied by drill bit, can increase productivity and life of drill.

“Almost as if it was a self-driving car, you have an algorithm that can interpret the data autonomously,” Mr. Sebregts said.

With AI, geologists’ time will be saved and they can supervise more wells and handle new problems, Shell said. Two applications of predictive maintenance build by the platform are reaching production stage:

First one covers upstream tools for production of coal seam gas in Australia.

Second one assists in identifying irregularities in downstream valves.

Using statistics to forecast the failure time of equipment, Shell will be able to maintain it before it actually breaks. Doing so, Shell can avoid sudden shutdown of its assets, and it will increase efficiency and decrease the costs.

Shell and other chief oil producers want to cut costs, increase production and effectively manage assets, by gaining from data coming out of both field equipment and corporate systems. This month, Chevron Corp. introduced a way to forecast maintenance problems in its refineries and oil fields.

Shell is also looking at other projects such as employing video analytics and image recognition to notify gas station supervisors when a customer does something dangerous e.g. smoking near the pump.  Shell expects that with AI platform, the data flowing across its business can be used to build a more influential decision-making tool.

The company has been using Azure for many years now, but it has only recently build-up its use. Application development platform is delivered by C3 IoT in addition to helping software apps for applications like fraud detection, predictive maintenance, and supply chain optimization. In Shell’s circumstance, the platform is in command of Microsoft’s Azure cloud and similar services.

Shell has made an agreement to operate C3 IoT on Azure for three years.

“When you look at enterprises like Shell, it’s really (about) how do you generate insights from that data,” Rohan Kumar, Azure Data’s corporate vice president for at Microsoft, said.